Online Issues
<< All Back-issues
<< This Issue's Table of Contents
ILAR Journal V40(4) 1999
Animal Models of Inflammation
Boris J. Czerrnak, Hans P. Friedl, and Peter A. Ward
| Boris J. Czermak, M.D., is House Officer, Department of Trauma Surgery, University of Freiburg School of Medicine, Freiburg/Breisgau, Germany; Hans P. Friedl, M.D., is Professor and Chairman, Department of Trauma Surgery, University of Freiburg; and Peter A. Ward, M.D., is Professor and Chairman, Department of Pathology, University of Michigan Medical School, Ann Arbor, Michigan. |
Introduction
One of the hallmarks of inflammation is the extrava-sation of neutrophils. The recruitment of neutro-phils from the bloodstream involves a sophisticated and interwoven chain of events. After generation of early response proinflammatory cytokines like interleukin (IL1)-1 or tumor necrosis factor-a (TNF(a1), adhesion molecules (such as E-selectin and ICAM-1) are expressed on surfaces of vascular endothelial cells. These molecules are designed to react with their respective "counter receptors" on neutrophils (such as complex oligosaccharides and b2 integrins) to cause margination, rolling, and adhesion of neutrophils along endothelial surfaces and finally, transmigration of neutro-phils through reversibly opened intercellular junctions in the endothelium. Once in the extravascular space, the directed migration of neutrophils depends on movement toward a chemical gradient of chemotactic factors generated within the inflamed tissue. Besides strong chemotactic factors like anaphylatoxins generated from the complement system (such as C5a), a major role for the process of mobilizing neutro-phiIs from the blood stream has been found for the family of IL-8-related chemoattractant cytokines (chemokines). There has been steady interest in this class of cytokines for almost 15 yr, but findings within the last several years--that the AIDS virus utilizes chemokine receptors to access cells--have intensified the research in this field (reviewed in Premack and Schall 1996).
Subclasses of chemokines are classified by the constellation of conserved cysteine residues. In the CC group of chemokines, the first two conserved cysteine residues are adjacent, and the CXC chemokines have one amino acid separating these conserved cysteine residues, in the CXXXC chemokine subfamily (fractalkine), three amino acids separate the two cysteine groups. The last group, the C chemokines (lymphotactin), is characterized by the absence of the first and third of the four conserved cysteine residues. Different subclasses of chemokines have been found to possess chemo-tactic activity for different cell types. CXC chemokines are strongly chemotactic for neutrophils, whereas CC chemokines principally attract cells of the monocyte/macrophage and lymphocyte lineage, although within individual chemokine clusters less selectivity and crossover activity can be found. Chemokines are produced by a variety of cell types. In the lung, macrophages, epithelial and endothelial cells, and fibro-blasts have been shown to be capable of synthesizing chemokines. This is also reflected by the presence of chemokines, especially IL-8, in bronchoalveolar lavage fluids of patients with adult respiratory distress syndrome (Miller and others 1992). Such findings have been derived from many experimental models dealing with the role of chemokines during lung inflammation. There are several well-characterized models of local or systemic inflammatory responses in rodents. Furthermore, rodent lung tissue is used as a source for a variety of cell lines utilized for in vitro experiments. In addition, rodents offer the possibility of genetic engineering with deletion of selected genes since extensive mapping of the mouse and rat genome has been done.
CXC Chemokines in Acute Lung Inflammatory Reactions in Rodent Models
For several of the human chemokines, there are no structurally identical chemokines in rodents; however, functionally equivalent chemokine homologues exist. For example, no molecule with a high degree of homology to human interleukin IL-8 has been found in either rats or mice. Instead, two structurally related CXC chemokines, macrophage inflammatory protein (MIP1)-2 and cytokine-induced neutrophil chemoattractant (CINC1), have been found to contain potent neutrophil chemoattractant and activating functions ascribed to human IL-8. These observations are reflected by the fact that rodents have IL-8-related receptors, which apparently react with MIP-2 and CINC (Cerretti and others 1993). MIP-2 was first identified in murine macrophages (Wolpe and others 1989) and found to be expressed at the mRNA level in several models of inflammation. Rat alveolar macrophages stimulated both in vitro and in vivo showed increased MIP-2 mRNA (Huang and others 1992). Rat neutrophils exposed to recombinant MIP-2 protein exhibited marked chemotactic activity; airway instillation of MIP-2 resulted in a significant influx of neutrophils into the airspace (Frevert and others 1995).
The role for MIP-2 and its close relative, CINC, was evaluated in vivo. In immunoglobulin G (IgG1) immune complex-induced lung injury in rats, both mRNA and protein for MIP-2 and CINC could be detected in lung in a time-dependent manner after initiation of the inflammatory reaction. Furthermore, specific blockade of either chemokine by antibody diminished lung injury by approximately 50% (as determined by vascular leakage of albumin) and reduced neutrophil influx into lung (as determined by neutrophil counts in bronchoalveolar lavage fluids) by more than 70% (Shanley and others 1997). Instillation of bacterial lipopolysaccharide (LPS1) into rat lungs also caused mRNA and protein expression of MIP-2. In the LPS-induced inflammatory model, a major amount of chemotactic activity present in bronchoalveolar lavage fluids was attributed to MIP-2, and the intratracheal blockade of this chemokine with antibody resulted in a sharp decrease in lung vascular leakage and pulmonary neutrophil influx (Schmal and others 1996). Because both MIP-2 and CINC are not expressed constitutively but are upregulated during acute inflammatory responses, we decided to study the mechanisms responsible for regulation of chemokine expression. In the immune complex lung inflammatory model, the inflammatory cascade is thought to be initiated by activated tissue macrophages. Some of the earliest events appear to be the local generation of complement activation products such as C5a and the release of the early response cytokines, TNFa and IL-1. We investigated both in vitro and in vivo the role of these early mediators by blockade of TNFa and C5a with specific antibodies. Rat alveolar macrophages were stimulated with IgG immune complexes in vitro in the presence of either anti-TNFa or anti-C5a, and chemokine content was assessed in cell culture supematant fluids. This approach was used in view of the evidence that alveolar macrophages generate both TNFa and C5a (Czermak and others 1999). In vitro production of MIP-2 and CINC by IgG immune complex-stimulated macrophages in the presence of either anti-TNFa or anti-C5a was reduced by 30 to 60% (Table 1). Intratracheal instillation of anti-TNFa or anti-rat C5a at the time of intrapulmonary IgG immune complex deposition also resulted in a similar marked decrease (30 to 56%) in both MIP-2 and CINC in bronchoalveolar lavage fluids (Table 2). Thus, neutrophil attractant CXC chemokines play an important role in acute inflammatory responses and occupy an early position in the chain of inflammatory events, suggesting that they may be targets for antiinflammatory interventions by direct or indirect blockade.
CC Chemokines in Acute Lung Inflammatory Reactions in Rodent Models
The CC chemokine subfamily is typified by MIP-1 and monocyte chemoattractant protein- 1 (MCP- 11). Both possess chemotactic activity for monocytes, although MIP-1 also has modest neutrophil chemokinetic activity (Wolpe and Cerami 1989). For MIP-1, two separate components, MIP-1a and MIP-1b, have been described. A dose-dependent increase in MIP-1a mRNA expression was demonstrated in rat alveolar macrophages stimulated in vitro with bacterial endotoxin (Shi and others 1995). In acute lung injury models induced by intratracheal instillation of LPS or intraalveolar deposition of IgG immune complexes, MIP-1a mRNA was found to be upregulated as a function of time, peaking at 6 hr after induction of injury. To assess whether appearance of MIP-1a was related to a proinflammatory role for this chemokine, rats were treated intratracheally with antibody to murine MIP-1a. The blockade of MIP-1a resulted in a distinct reduction in the permeability index (circa 40%) in both models. Neutro-phil influx into the alveolar space was greatly ameliorated in the IgG immune complex-induced lung injury and almost completely abolished in LPS-induced alveolitis (Shanley and others 1995). In both models, TNFa levels in bronchoalveolar lavage fluids were substantially diminished, a requirement that has also been demonstrated for upregulation of ICAM-1. These findings suggest a role for MIP-1a as an autocrine stimulator of TNFa generation and downstream mediators such as chemokines. The interaction between TNFa and MIP-1a has been further elucidated by recent results, which show a requirement for TNFa in the in vitro generation of MIP- 1a by macrophages, a finding that could also be reproduced in vivo for pulmonary MIP-1a production in the rat lung injury model induced by IgG immune complexes (Czermak and others 1999). Thus, there appears to exist a positive autocrine feedback mechanism involving TNFa, which (together with C5a) enhances macrophage production of MIP-1a.
The importance of MIP-1a as mediator of virus-induced inflammation has been demonstrated by the finding that mice whose gene encoding this chemokine was disrupted ("knockout mice") were resistant to Coxsackie virus-induced myo-carditis. These mice also displayed reduced influenza virus-induced lung inflammation and delayed clearance of the virus (Cook and others 1995).
MCP-1 is a chemotactic and activating CC chemokine for mononuclear leukocytes. Its role was first described as a potent mediator for fibroblast activation and proliferation. Selective monocyte attracting function was later found and has been linked to progress of the inflammatory component of vascular diseases such as atherosclerosis. In the neutrophil-independent rat model of immunoglobulin A immune complex-induced alveolitis (which is a macrophage-dependent but neutrophil-independent process), treatment with blocking antibody to MCP-1 led to a marked reduction in lung injury as measured by pulmonary vascular permeability, alveolar hemorrhage, and pulmonary monocyte/macrophage recruitment (Jones and others 1992).
Because generation of CXC chemokines has been shown to require TNFa and C5a for full expression (see above), we investigated the dependency of CC chemokine production on these early upstream mediators. We stimulated alveolar macrophages in vitro with IgG immune complexes in the presence or absence of blocking antibodies to either C5a or TNFa and assessed the content of MIP-1a, MIP-1b, and MCP-1 in cell culture supernatant fluids. The copresence with stimulated macrophages of anti-TNFa or anti-C5a reduced the levels of these chemokines by nearly 40% compared with controls (Table 1). During in vivo experiments, we induced lung injury in rats by intraalveolar deposition of IgG immune complexes, and we quantitated CC chemokines in bronchoalveolar lavage fluids at 4 hr. When we treated the animals intratracheally with either anti-TNFa or anti-C5a, the intrapulmonary generation of MIP-1a and MIP- 1 decreased 40 to 50% (Table 2). Interestingly, the effect of blocking C5a or TNFa was not as pronounced for intrapulmonary production of MCP-1, where we noted smaller reductions (15 to 25%) of this chemokine.
Thus, a somewhat different pattern of mediator-dependent regulation was demonstrated for CXC and CC chemokines. It is not known whether C5a and TNFa exert their controlling functions on chemokine generation by similar or different mechanisms. There are reports about differences in intra-cellular signal transduction pathways employed after stimulation with either TNFa or C5a. For example, C5a-induced endothelial cell production of superoxide involves G protein and protein kinase C pathways, whereas no involvement of these signal transduction pathways has been demonstrated for superoxide production after stimulation with TNFcz (Murphy and others 1992). In contrast, upregulation of the nuclear transcription factor kappa B (NFkB) was induced by intratracheal instillation of TNFa but not C5a (Lentsch and others 1998), confirming that activation events caused by TNFct or C5a operate via different intracellular signal tranduction pathways. Because the blockade of either C5a or TNFa failed to inhibit chemokine production completely, other regulatory mechanisms or upstream inflammatory events might be involved in this process.
Conclusions
The important role of leukocyte influx into the site of inflammation is well documented, and the mechanisms responsible for recruitment of these cells are an integral part of host defense reactions. The use of animal models has greatly increased our understanding of the role of chemokines in such acute inflammatory reactions. Because the lung contains a defined number of cell populations and has been a target for intensive study of the inflammatory response, this organ is a particularly interesting and revealing subject for in vivo inflammatory research. Insights gained from the use of lung injury models in rodents as well as from targeted disruption or overexpression of the various cytokine and chemokine genes will be valuable in defining more completely the role of these genes in histological system. This increased knowledge should lead to more useful and effective intervention in many clinical conditions involving the inflammatory response.
1 Abbreviations used in this paper: CINC, cytokine-induced neutrophil chemoattractant; lgG, imnmnoglobulin G; IL, interleukin; LPS, lipopolysaccharide; MCP- 1, monocyte chemoattractant protein- 1; MIP, macrophage inflammatory protein; TNFa, tumor necrosis factor a.
References
Cerretti DP, Kozlosky CJ, Vanden Bos T, Nelson N, Gearing DP, Beckmann MP. 1993. Molecular characterization of receptors for human interleukin-8, GRO/melanoma growth-stimulatory activity, and neutro-phil activating peptide-2. Mol Immunol 30:359-367.
Cook DN, Beck MA, Coffman TM, Kirby SL, Sheridan JF, PragneU IB, Smithies O. 1995. Requirement of MiP- 1 cc for an inflammatory response to viral infection. Science 269:1583-1585.
Czermak B J, Sarma V, Bless NM, Schmal, Peter Friedl Hans, Ward PA. 1999. In vitro and in vivo dependency of chemokine generation on C5a and TNFa. J Immunol 162:2321-2325.
Frevert CW, Farone A, Danaee H, Paulauskis JD, Kobzik L. 1995. Functional characterization of rat chemokine MIP-2. Inflammation 19:133-142.
Huang S J, Paulauskis D, Godleski J J, Kobzik L. 1992. Expression of macrophage inflammatory protein 2 and KC mRNA in pulmonary inflammation. Am J Pathol 141:981-988.
Jones ML, Mulligan MS, Flory CM, Ward PA, Warren JS. 1992. Potential role of monocyte chemoattractant protein 1/JE in monocyte/macrophage-dependent IgA immune complex alveolitis in the rat. J Immunol 149:2147-2154.
Lentsch AB, Czermak BJ, Bless NM, Ward PA. 1998. NFkB activation during IgG immune complex-induced lung injury. Requirements for TNFa and IL-1b but not complement. Am J Pathol 152:1327-1336.
Miller EJ, Cohen AB, Nagao S, Griffith D, Maunder RJ, Martin TR, Weiner-Kronish JP, Sticherling M, Christophers E, Matthay MA. 1992. Elevated levels of NAP-1/interleukin-8 are present in the airspaces of patients with adult respiratory distress syndrome and are associated with increased mortality. Am Rev Respit Dis 146:427-432.
Murphy HS, Shayman JA, Till GO, Maroughui M, Owens CB, Ryan US, Ward PA. 1992. Superoxide responses of endothelial cells to C5a and TNFa: Divergent signal transduction pathways. Am J Physio1263 :L51-L59.
Premack BA, Schall TJ. 1996. Chemokine receptors: Gateways to inflammation and infection. Nature Med 2:1174-1178.
Schmal H, Shanley TP, Jones ML, Friedl HP, Ward PA. 1996. Role for MIP-2 in lipopolysaccharide-induced lung injury in rats. J Immunol 156:1963-1972.
Shanley TP, Schmal H, Friedl HP, Ward PA. 1995. Role of macrophage inflammatory protein-1a (MIP-1a) in acute lung injury in rats. J Immunol 154:4793-4802.
Shanley TP, Schmal H, Warner RL, Schmid E, Friedl HP, Ward PA. 1997. Requirement for C×C chemokines MIP-2 and CINC in lgG immune complex-induced lung injury. J Immunol 158:3439-3448.
Shi MM., Godleski JJ, Paulauskis JD. 1995. Molecular cloning and post-transcriptional regulation of macrophage inflammatory protein- 1 alpha in alveolar macrophages. Blochem Biophys Res Commun 211:289-295.
Wolpe SD, Cerami A. 1989. Macrophage inflammatory protein 1 and 2: Members of a novel superfamily of cytokines. FASEB J 3:2565-2573.
Wolpe SD, Sherry B, Juers D, Davatelis G, Yurt RW, Cerami A. 1989. Identification and characterization of macrophage inflammatory protein 2. Proc Natl Acad Sci U S A 86:612-616.
Table 1 In vitro requirements for tumor necrosis factor a (TNFa) and C5a in chemokine generation by rat alveolar macrophages stimulated with immunoglobulin G immune complexes (ICs)
| CXC | CC | ||||
| MIPa-2 | CINCa | MIP-1a | MIP-1b | MCPa- 1 | |
| Untreated (ng/mL)b | 34.2 ± 1.94 | 76.6 ± 7.2 | < 1 | 15.7 ± 3.19 | 9.65 ± 0.71 |
| lC (ng/mL)c | 90.8 ± 5.33 | 703 ± 51.8 | 392 ± 38.5 | 280 ± 8.22 | 19.4 ± 0.32 |
| lC + anti-TNFa | -38.5 % | -30 % | -39.7 % | -41.3% | -43.6 % |
| lC + anti-C5a | -60.9 % | -32.8 % | -40.2 % | -46.3 % | -42 % |
bThese numbers were used to establish the 0% value.
c These numbers were used to establish the 100% value.
Table 2 In vivo requirement for tumor necrosis factor a (TNFa) and C5a in chemokine expression in inflamed rat lung after immunoglobulin G immune complex (lC) deposition
| MIPa-2 | CINCa | MIP-1a | MIP-1b | MCPa-1 | |
| Untreated (ng/mL)b | 0.55 ± 0.63 | < 1 | < 1 | 8.91± 3.2 | < 1 |
| lC (ng/mL)c | 42.4 ± 2.19 | 1291 ± 45 | 1764 ± 111 | 448 ± 13.8 | 14.4 ± 0.46 |
| lC + anti-TNFa | -31.1% | -27.8 % | -37.3 % | -36.5 % | -26.3 % |
| lC + anti-C5a | -43.6% | -48.6 % | -56.2% | -50.7 % | -17.4 % |
b These numbers were used to establish the 0% value.
c These numbers were used to establish the 100% value.
Copyright © 2008. National Academy of Sciences.
All rights reserved.
500 Fifth St. N.W., Washington, D.C. 20001.
Terms of Use and Privacy Statement